Selectively Introducing Acid Sites in Different Confined Positions in

Publication Date (Web): July 10, 2018 ... sites is determined on the basis of the product distribution of 1-hexene cracking as test reaction. ... 0 (0...
1 downloads 0 Views 2MB Size
Subscriber access provided by UNIV OF DURHAM

Article

Selectively Introducing Acid Sites in Different Confined Positions in ZSM-5 and its Catalytic Implications Chengeng Li, Alejandro Vidal-Moya, Pablo J. Miguel, JIRI DEDECEK, Mercedes Boronat, and Avelino Corma ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.8b02112 • Publication Date (Web): 10 Jul 2018 Downloaded from http://pubs.acs.org on July 14, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 11 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Selectively Introducing Acid Sites in Different Confined Positions in ZSM-5 and its Catalytic Implications Chengeng Li,† Alejandro Vidal-Moya,† Pablo J. Miguel,§ Jiri Dedecek,‡ Mercedes Boronat†,* and Avelino Corma†,* †

Instituto de Tecnología Química, Universitat Politècnica de València, Consejo Superior de Investigaciones Cientificas, Av. de los Naranjos, s/n, 46022 Valencia, Spain § Departamento de Ingeniería Química, Universitat de València, Av. de la Universitat, s/n, 46100 Burjassot, Spain ‡ J. Heyrovsky Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, Dolejskova 3, CZ-182 23 Prague 8, Czech Republic Supporting Information

ABSTRACT: Controlling the location of acid sites in zeolites can have a great impact on catalysis. In this work we face the objective of directing the location of Al into the 10R channels of ZSM-5 by taking advantage of the structural preference of B to occupy certain positions at the channels intersections, as suggested by theoretical calculations. The synthesis of B-Al-ZSM-5 zeolites with variable Si/Al and Si/B ratio, followed by B removal in a post synthesis treatment, produces ZSM-5 samples enriched in Al occupying positions at 10R channels. The location of the acid sites is determined on the basis of the product distribution of 1-hexene cracking as test reaction. The higher selectivity to propene and lower C4=/C3= ratio in the samples synthesized with B and subsequently deboronated can be related to a larger concentration of acid sites in 10R channels, where monomolecular cracking occurs. Finally, several ZSM-5 samples have been tested in the methanol to propene reaction, and those synthesized through the B assisted method show longer catalytic lifetime, higher propene yield and lower yield of alkanes and aromatics.

ed straight (10R-str) and sinusoidal (10R-sin) ten-membered ring channels of ⁓ 5.0 Å diameter, that intersect forming larger void spaces of ⁓ 7.0 Å diameter (see Figure 1). The Brönsted acid sites resulting from isomorphic substitution of Al by Si in ZSM-5 can point towards the channels or towards the open space generated at the channels intersection depending on Al location. Despite the fact that both types of acid sites can interact with a given reactant molecule, the molecular confinement and therefore the enthalpic and entropic effects on the adsorbed reactant and on the transition state in the two environments will not be necessarily the same.7-13

KEYWORDS: Zeolites, DFT, Al-siting, Boron ZSM-5, MTO, catalytic cracking.

INTRODUCTION Controlling the location of acid sites in zeolites can have a great impact on their catalytic behavior. The Brönsted acidity in zeolites is associated to protons compensating the negative charge generated by the presence of framework Al in tetrahedral T sites. The complexity of the crystal structure in zeolite frameworks generates multiple topologically distinguishable T sites that, when occupied by Al, originate Brönsted acid sites with the protons facing different cavities or channels. The intrinsic acid strength of Brönsted sites associated to Al located at different T positions is similar when described rigorously by deprotonation energies (DPE), 1-5 but their catalytic behavior in terms of activity and selectivity can differ with the location, due to the influence of confinement effects on intermediates and transition states.6-10 A paradigmatic case in where the control of the acid site distribu tion can have important practical implications is the MFI framework, whose aluminosilicate form is the medium pore ZSM-5 zeolite. The channel system of MFI is composed by interconnect-

Figure 1. Crystallographic T positions in MFI framework. ZSM-5 is one of the two zeolites used in the industrial transformation of methanol to olefines (MTO), the other one being SAPO-34 with the CHA structure. The MTO process is an efficient way to obtain light olefins, such as ethene and propene, by successive methylation and cracking of the so-called hydrocarbon pool (HP) organic intermediates that act as co-catalysts of the reaction.14-16 The detailed reaction mechanism, product distribution and catalyst deactivation are closely related to the zeolite topology and pore architecture. In the larger channels and cavities of BEA and CHA catalysts, polymethylbenzenes (PMB) are the main HP species, and the side-chain route generating both ethene and propene is the most probable pathway. In MFI, a dual-cycle mechanism has been proposed, according to which ethene is formed via the aromatics-based HP mechanism while propene and

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

higher alkenes are formed through methylation/cracking of C3+ alkene intermediates.14,15,17-19 The suppression of the aromaticsbased cycle in H-ZSM-22 zeolite containing an unidirectional system of non-intersecting straight 10R channels20 suggests that, in H-ZSM-5, the polymethylbenzene HP species are formed in the channels intersections, while the alkene-based cycle is active within the 10R channels. Therefore, it is inferred that the propene/ethene selectivity in the MTO process, as well as the amount of aromatics and alkanes resulting from bimolecular hydrogen transfer processes preferentially occuring at the channels intersections, could be modified by controlling the location of the Brönsted acid sites within the MFI framework. Researchers have attempted to tune acid site location in FER,21,22 CHA,23,24 RTH25 and MFI26-32 zeolite frameworks by developing new synthetic strategies that combine different aluminium or silicon sources, organic structure directing agents (OSDAs) and inorganic compensating cations. Despite changes in catalytic performance associated to the synthetic route employed have been reported several times, only in some cases the origin of the catalytic diversity could be identified, and related either to the presence of Al at particular cavities or channels21,22 or to its nature as isolated or paired Al centres.24 The unambiguous identification of the Al siting in the 12 distinct T sites of H-ZSM-5 zeolite is particularly challenging, and has been attempted by means of 27AlMAS-NMR, UV-Vis of Co2+, chemical adsorption and catalytic experiments.26-32 Deconvolution of the 27Al-MAS-NMR spectra of ZSM-5 into different peaks has allowed to demonstrate that Al siting in this zeolite varies with the synthesis conditions, and that several different T sites are occupied by Al even in samples with high Si/Al ratio.33-35 However, while different deconvolution patterns are indicative of a different Al distribution within the MFI framework, a clear assignation of the peaks to particular positions is not established yet. Another aspect considered in relation to Al distibution is the proximity between Al atoms, and their local arrangement as isolated, paired - defined as framework Al atoms separated by either one or two Si atoms [Al-O(-Si-O)xAl, x=1,2]- or unpaired species, which are separated by more than three Si atoms but are close enough to compesate the charge of a divalent cation. The concentration of Al in each of the three arrangements can be estimated by combining Co2+ and Na+ ionexchanging capacity, and the characterization of the paired species as α, β and γ can be determined by deconvolution of the diffuse reflectance UV-Vis spectra of dehydrated Co2+ exchanged samples.26,27,30,36 Another strategy to control the Al distribution within a given zeolite framework is based on isomorphous substitution with a second trivalent heteroatom, usually boron, that might preferentially compete with Al for some T sites. The succesful regulation of the Al siting through competitive incorporation of B has been recently achieved in H-MCM-22 zeolite. The preferential location of B in T positions facing the surface pockets and supercages of the MWW structure, prone to carbonaceous depositions, concentrates the Brönsted acid sites associated to Al into the 10R sinusoidal channels, where the alkene-based cycle is sterically favored. As a consequence, the modified catalyst exhibits a high selectivity to propene and butenes, and an improved long-term stability.37,38 There are some examples in the literature reporting an enhanced stability against deactivation by coke formation in B-modified ZSM-5 catalysts, which has been attributed to an increase in the amount of weak acid sites associated to B.39-42 However, no clear trends in the propene/ethene selectivity have been observed, and no direct relationship between B or Al siting and catalytic performance has been established. In this work we use the competitive Al and B incorporation in the ZSM-5 framework as starting working hypothesis to preferentially locate Al atoms and the associated Brönsted acid sites in the 10R channels of ZSM-5, with the final objective of increasing the

Page 2 of 11

selectivity to propene, reducing the amount of aromatics and alkanes, and improving the catalyst stability. To achieve this goal, we have first studied by Density Functional Theory (DFT) the preferred location, if there is any, of Al and B in the MFI structure. Then, Al-B-ZSM-5 zeolites with different Si/Al and Si/B ratios have been synthesized and characterized, and in a second step B has been selectively removed by post synthesis treatments. To determine the relative population of Brönsted acid sites associated to the remaining Al in the 10R channels or at the channels intersections we have used the product selectivity obtained in a test reaction, 1-hexene cracking, that undergoes different mono- and bimolecular reactions depending on the space available around the active sites. Comparison with the results obtained using a unidimensional 10R pore zeolite (Theta-1) leads to the conclusion that the method proposed allows to modify the relative amount of Al atoms, and consequently of acid sites, within the channels versus channels intersections, and that it is possible to prepare, on those bases, improved catalysts for the conversion of methanol to propene. RESULTS AND DISCUSSION DFT study of Al and B location in ZSM-5 The hypothesis that B and Al tend to occupy different T positions in ZSM-5 was investigated by means of DFT calculations (see Computational Details in the Supporting Information). Among the twelve non-equivalent tetrahedral T sites present in the orthorrombic structure of ZSM-5, T4 and T10 are located in the sinusoidal 10R-sin channel, T8 and T11 are in the straight 10R-str channel, and the rest (T1, T2, T3, T5, T6, T7, T9 and T12) are at the intersection between the 10R-sin and 10R-str channels. Previous quantum chemical calculations agree that the difference in energy between the most and least stable location of Al is not larger than 9 kcal/mol.11,43,44 In a first step, six different systems with chemical composition AlxSi96-xO192 and BxSi96-xO192, with x = 1, 2 and 4 were generated to model ZSM-5 zeolite with different Si/Al and Si/B ratios, and twelve different distributions corresponding to placing the Al or B atoms in the same crystallographic T position, from T1 to T12, were considered for each chemical composition. The negative charge generated in the framework by the presence of Al an B was compensated by tetraethyl ammonium cations in a tg.tg conformation occluded at the channels intersections.45 The relative stabilities of all distributions with respect to the most stable one are summarized in Table 1. Table 1. Relative stability of Al and B distributions (in kcal/mol) of Al-ZSM-5 and B-ZSM-5 zeolite models with different Si/Al and Si/B ratios. Si/Al Si/B 95 47 23 95 47 23 T1 0.0 1.2 1.6 3.4 3.0 9.0 T2 3.1 6.4 2.6 4.9 9.3 7.4 T3 2.6 5.7 0.4 6.4 11.4 13.2 T4 1.4 0.6 2.8 5.9 7.8 22.5 T5 1.7 0.9 5.4 0.7 0.0 6.7 T6 4.0 8.0 6.4 2.1 2.1 0.0 T7 3.3 8.3 16.3 3.4 6.6 17.4 T8 1.9 3.6 4.6 5.4 9.5 22.0 T9 3.1 5.8 5.3 3.5 6.1 6.1 T10 1.5 4.5 5.4 0.0 0.5 2.2 T11 1.3 0.0 0.9 1.9 1.4 7.8 T12 0.1 0.1 0.0 3.5 5.1 8.6 In agreement with previous studies,11,43,44 the energy values obtained for the Al-containing models are quite similar, within 4 kcal/mol at high Si/Al ratio and within 8 kcal/mol at most at higher Al content, with the only exception of T7 position that is clearly unstable at Si/Al = 23 (see Table 1 and Figures 2 and S1).

ACS Paragon Plus Environment

Page 3 of 11 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

This result is in accordance with the proposal that Al siting in ZSM-5 depends on the synthesis procedure,33-35 because there is not any clear thermodynamic preference for occupying any given position in the framework. Indeed, at the highest Al content considered, there are six different positions whose relative stability differs by less than 3 kcal/mol: T1, T2, T3 and T12 at the channels intersections and T4 and T11 within the 10R channels. In contrast, in the B-containing models, differences in stability larger than 6 kcal/mol are already obtained at high Si/B ratio, that significantly increase with the B content. Thus, only T5, T6, T10 and T11 are within 3 kcal/mol at low and medium B content. Increasing the Si/B ratio to 23 results in a clear destabilization of most distributions, and the results in Table 1 and Figures 2 and S1 indicate that B atoms exhibit a clear thermodynamic preference to occupy T6 position at the channels intersection followed by T10 in the sinusoidal 10R-sin channel. This preferential siting of B at T6 and T10 is maintained in mixed Al-B-ZSM-5 models in which two Al and two B atoms are placed at the same T position (see Table 2). Table 2. Relative stability of Al and B distributions (in kcal/mol) of Al-B-ZSM-5 zeolite models with Si/(Al+B) = 23 and Al/B = 1. Al B Erel Al B Erel Al B Erel T1 T1 3.6 T4 T5 5.3 T11 T6 3.2 T2 T2 6.5 T4 T6 1.7 T11 T10 4.6 T3 T3 6.8 T4 T10 1.9 T12 T5 5.9 T4 T4 8.7 T4 T11 7.9 T12 T6 1.2 T5 T5 5.1 T5 T6 10.5 T12 T10 3.3 T6 T6 3.6 T5 T10 4.8 T12 T11 7.3 T7 T7 15.9 T5 T11 7.2 T6 T4 16.2 T8 T8 13.8 T8 T5 10.7 T6 T5 11.9 T9 T9 7.5 T8 T6 1.4 T6 T8 13.3 T10 T10 3.2 T8 T10 7.9 T6 T10 3.6 T11 T11 7.0 T8 T11 9.0 T6 T11 8.2 T12 T12 8.4 T10 T5 9.7 T6 T12 9.8 T1 T5 5.5 T10 T6 9.5 T2 T6 0.0 T1 T6 7.2 T10 T11 8.5 T3 T6 3.2 T1 T10 1.3 T10 T4 11.9 T7 T6 10.7 T1 T11 4.1 T11 T5 9.5 T9 T6 3.8 To try to understand the origin of the higher stability of B substitution at certain T positions, some geometrical parameters like the average Si-Si bond length in the pure silica material or the distance between the T atom and the N atom of the quaternary ammonium cation are summarized in Table S1 together with the relative stability of each Al and B distribution at high Si/Al and Si/B ratios. While no clear trend is observed for Al distribution, the plot in Figure S2 suggests that B is more stable at those T sites with shorter Si-Si distances, in line with the smaller size of the BO4 unit as compared to AlO4. Then, other possible distributions of Al and B in which the two heteroatoms occupy different T positions were explored, with B being preferentially placed at T5, T6, T10 and T11 due to the significant differences in stability obtained for substitution of this heteroatom and the large computational effort required to include all possibilities. Data in Table 2 confirm the trend that the most stable systems (within ⁓ 4 kcal/mol) contain B in either T6 or T10, irrespectively of the location of Al. Indeed, exchanging the Al and B positions in the most stable configurations always led to a significant destabilization of the system, as for instance T4-T6, T4-T10, or T6-T8 (see Table 2). It should also be noticed that three out of the six most stable distributions, whose relative energy differs by less than 3 kcal/mol, correspond to Al siting at T4, T8 and T11, that is, within the 10R channels (see Table 2 and Figure 2). Comparison of the plots in Figure 2a and 2c suggests that the presence of B increases the proportion of stable configurations containing Al in the 10R channels. It can then be conclud-

ed from the DFT calculations that, from a thermodynamic point of view, Al has little tendency to occupy preferentially any of the 12 T positions in ZSM-5 zeolite, in agreement with previous reports.11,43,44 On the other hand, B prefers to occupy T6 and T10 sites at or close to the channels intersection rather than any other position and, consequently, the proportion of T sites within the channels occupied by Al increases.

Figure 2. Relative energy (in kcal/mol) with respect to the most stable distribution in a) Al-ZSM5 with Si/Al=23, b) B-ZSM5 with Si/B=23, and c) Al-B-ZSM-5 with Si/(Al+B)=23 and Al/B=1.Full and open symbols indicate location of Al and B, respectively. T sites at channels intersections are plotted in blue, orange squares and circles correspond to T sites within the 10R-sin and 10R-str channels, respectively. Some T sites are indicated by numbers. Synthesis and Characterization of Al-B-ZSM-5 zeolites It can therefore be envisioned that, by synthesizing Al-B-ZSM-5 zeolites with a Si/(Al+B) ratio of ⁓ 23 and Al/B ratio of ⁓ 1, and selectively removing the B with post-synthesis treatments, we could obtain Al-ZSM-5 samples with a higher proportion of Al in the channels than in the case of conventionally prepared Al-ZSM5 zeolite. To confirm this hypothesis, Al-ZSM-5 and B-ZSM-5 samples with different Si/Al and Si/B ratios, as well as the corresponding Al-B-ZSM-5 samples with different B/Al ratios were prepared (Table 3). The Si/(Al+B) ratio in the B-containing gels ranged from 11.5 to 33.3. Notice that the Ttotal/OSDA ratios (Table 4) in one unit cell in all the as-synthesized samples were calculated to be near 24, based on the results of elemental analysis. This is, as well, the amount of available positive charges for TIII heteroatoms to be located, limiting the minimum Si/(Al+B) in the resultant product to be 24. As a result, the cases with Si/(Al+B) lower than 24 should force competitive incorporation of Al and B in the available sites. Data in Table 3 indicate that, in these cases (samples labelled -B2), all Al present in the synthesis gel is incorporated into the zeolite framework while only a fraction of the B initially present in the gel is finally located in the zeolite structure. Crystallinity, chemical analysis and textural properties of the SiAl, Si-B and Si-Al-B ZSM-5 zeolites before and after deboronation are given in Figure 3 and Table 4. The XRD patterns of all synthesized ZSM-5 samples showed the characteristic diffraction peaks of the MFI topology (Figure 3A). After the deboronation and calcination process the samples preserved the crystallinity (Figure 3B), indicating that the removal of most of the B (Table 4) does not cause notable destruction of the framework structure. This conclusion is supported by the observation that the mi-

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

cropore volume of the zeolite samples remains practically the same after the deboronation treatment. It is also important to notice that no changes in the Al content of the Si-Al-B samples was observed due to the above postsynthesis treatments (Table 3).

Z5-100-B2 11.78 23.3 a measured by elemental analysis

Page 4 of 11 388

0.17

41

Figure 4. SEM images of ZSM-5 samples (a) Z5-50 (b) Z5-50-B1 (c) Z5-50-B2 (d) Z5-75 (e) Z5-75-B1 (f) Z5-75-B2 (g) Z5-100 (h) Z5-100-B1 (i) Z5-100-B2 (j, k) Z5-50-B1 (l) Z5-pureB. The scale bar in the images represents a length of 200 nm.

Figure 3. XRD patterns of ZSM-5 samples in as-synthesized forms (A) and after deboronation process (B). Table 3. Chemical composition of different ZSM-5 samples. Composition gela As-synthesizeda Deboronatedb Sample Si/Al Si/B Si/(Al+B) Si/Al Si/B Si/(Al+B) Si/Al

Z5-pureB - 50 50 Z5-50 50 50 Z5-50-B1 50 50 25 Z5-50-B2 50 15 11.5 Z5-75 75 75 Z5-75-B1 75 50 30 Z5-75-B2 75 15 12.5 Z5-100 100 100 Z5-100-B1 100 50 33.3 Z5-100-B2 100 15 13.0 a measured by ICP-AES b measured by elemental analysis

43 44 47 58 60 72 86 87 90

45 49 49 67 40 77 39

45 43 23.2 24.0 58 31.7 25.7 86 40.8 27.2

44 46 48 58 61 70 86 90 92

Table 4. Characterization of different ZSM-5 samples. OSDA SBET Ttotal/ Vmicro Sample (wt%)a OSDA (m2/g) (cm3/g) Z5-pureB 11.79 23.2 382 0.17 Z5-50 11.47 24.0 401 0.17 Z5-50-B1 11.28 24.4 364 0.17 Z5-50-B2 11.34 24.3 400 0.16 Z5-75 11.57 23.7 378 0.16 Z5-75-B1 11.14 24.8 391 0.17 Z5-75-B2 11.62 23.6 395 0.17 Z5-100 11.45 24.0 375 0.16 Z5-100-B1 11.31 24.4 390 0.17

Si/B

107 82 96 117 64 149 78

Sext (m2/g) 47 53 49 83 41 61 44 43 38

Another key parameter to define and compare catalytic activity of different zeolite samples is the crystallite size, which has been measured here by FESEM. The results obtained (Figure 4) indicate that all ZSM-5 samples, regardless of their framework composition, are formed by nanosized crystals of 100-200 nm. Moreover, samples with medium and high Si/Al ratios (Z5-75 (Figure 4d-f) and Z5-100 series (Figure 4g-i)) have spherical morphology with clean external surface. Samples of the Z5-50 series with relatively higher Al content show also the presence of smaller crystallites (Figure 4a-c). The internal sections of the samples analyzed by FIB-SEM (Figure 4j, k) illustrate that, despite some aggregation of the nanoparticles occur, the crystals were well defined with sizes between 100 and 200 nm, being the pattern very similar for all the samples. Therefore we will be able to compare the catalytic activity of the samples on the bases of their framework composition and the way of achieving this composition for a constant zeolite crystallite size. The acid properties of the deboronated ZSM-5 samples were characterized by NH3-TPD and FT-IR combined with pyridine adsorption/desorption measurements (see Figure S3, S4 and Table S2 in the Supporting Information). The total amount of acid sites correlates with the Si/Al ratio of the samples, and the absence of pyridine adsorption in a pure B-ZSM-5 zeolite prepared for comparison purposes confirms that residual B does not significantly contributes to the measured acidity after removal of B. 27 Al MAS NMR spectra (Figure 5) show the presence of an intense peak at ~ 55 ppm assigned to tetrahedrally coordinated Al in all samples, while the band corresponding to extraframework Al at ~ 0 ppm was not observed in any case. In order to obtain information about the Al distribution in each sample, the broad signal ranging from 45 to 65 ppm was deconvoluted into five different peaks at 52, 53, 54, 56 and 58 ppm (see the 27Al 3QMAS NMR spectrum of the Z5-50 and Z5-50-B samples in Figure S5) following the procedure previously described by Yokoi et al.28 The curve fittings and relative peak areas of the 27 Al MAS NMR spectra of the different ZSM-5 samples are given in Figure 6 and Table 5. The relative concentration of

ACS Paragon Plus Environment

Page 5 of 11 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

the five peaks differs noticeably from one sample to another, with the only exception of the two samples with higher Si/Al ratio, Z5-100 and Z5-100-B1, which are quite similar. However, there is not a clear trend regarding the changes in intensity associated to the presence of B in Z5-50 and Z5-75 series. It can therefore be concluded that there is a different distribution of Al atoms over the 12 non-equivalent tetrahedral T sites present in ZSM-5 in the samples with low Si/Al ratio caused by the competitive introduction of B in the framework, an effect that is not observed at high Si/Al ratio. However, it is not possible at this point to assign the peaks to particular T positions and therefore to identify the location of Al based only on 27Al MAS NMR spectra.

Figure 6. Curve fitting of the 27Al MAS NMR spectra of (A) Z550, (B) Z5-50-B1, (C) Z5-75, (D) Z5-75-B1 (E) Z5-100 and (F) Z5-100-B1. Table 5. Relative peak areas of the 27Al MAS NMR spectra of different ZSM-5 samples. Al(e) Al(d) Al(c) Al(b) Al(a) chemical shift (ppm) 58 56 54 53 52 Z5-50 6 33 18 30 13 Z5-50-B1 13 26 20 20 21 Z5-75 14 22 15 21 28 Z5-75-B1 12 30 13 28 17 Z5-100 6 36 19 24 15 Z5-100-B1 5 36 20 26 12 Table 6. Co2+-exchange experiment results of different ZSM-5 samples. Sample

Compositiona Si/Al

Si/B

Co-exchange experiment

Alisolb

Alpairc

Alαd

Alβd

Alγd

Z5-50 44 45% 55% 32% 65% 3% Z5-50-B1 46 82 30% 70% 45% 55%